How do the systems behind all our connected devices become smarter and more reliable? This is the issue that a major project at Malmö University is set to explore.

The Internet of Things (IoT) is about all our connected devices, from telephones and network cameras to wireless speakers and smartwatches.

If you have an alarm system that includes cameras and microphones to detect burglary, it can be safer if the analysis is done locally on a device in the home, instead of images and sound from the home being spread to the cloud.

Paul Davidsson

“We want the IoT systems to be as smart as possible, but we also want to be able to trust them. We want to understand what they do and know that they protect users' privacy, despite the fact that they collect an incredible amount of data,” says Paul Davidsson, professor of computer science at the University’s research centre, Internet of Things and People (IoTaP).

Davidsson is the project manager for the three-year project, which is funded by the Knowledge Foundation together with participating technology companies. The question is therefore, how to develop intelligent and credible IoT systems. To answer this, the researchers will investigate how artificial intelligence can be used in the best way. They will also look at issues concerning how data can be processed closer to the user, so-called edge computing, instead of being uploaded the cloud.

“If you have an alarm system that includes cameras and microphones to detect burglary, it can be safer if the analysis is done locally on a device in the home, instead of images and sound from the home being spread to the cloud,” says Davidsson.

Doing data calculations locally and not in the cloud reduces the amount of data communication which consumes a lot of energy. The research project thus also includes a sustainability and environmental aspect.

The research questions have been developed together with the companies and will be investigated on the basis of four different application areas: health, energy, buildings, and monitoring systems. The goal is to be able to develop more reliable and stable but also more intelligent and powerful systems for both the individual and society at large.

“The idea is that we will develop a general framework that developers of IoT systems can benefit from and can use,” says Davidsson.

The Knowledge Foundation's CEO Eva Schelin says that the overall scientific competence of the research group makes for good results.

“Participating companies are relevant to the project and are expected to benefit from the project. The research issues concerning system architecture, AI and data integrity are crucial for the development of the participating companies,” says Schelin.

Text: Magnus Jando & Adrian Grist